From pradermecker at yahoo.ca Tue Jul 1 11:46:51 2014 From: pradermecker at yahoo.ca (Pierre Radermecker) Date: Tue, 1 Jul 2014 04:46:51 -0700 Subject: [arch-haskell] xsounds.org deprecated or down ? In-Reply-To: References: <1404122612.27100.YahooMailNeo@web161506.mail.bf1.yahoo.com> Message-ID: <1404215211.59130.YahooMailNeo@web161503.mail.bf1.yahoo.com> Probably something wrong on my part, then. I suspect timeout issue. Thanks for your help On Monday, June 30, 2014 1:37:37 PM, Magnus Therning wrote: I'm currently out travelling so I'm hoping this is a temporary thing, or a question of configuration or your side ;) /M On 30 Jun 2014 05:03, "Pierre Radermecker" wrote: Hi, > > >`xsounds.org` has been down for a few days now. Is it still a correct pointer for [haskell-core]? ? > > > >Cheers, >_______________________________________________ >arch-haskell mailing list >arch-haskell at haskell.org >http://www.haskell.org/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bb010g at gmail.com Tue Jul 8 18:39:26 2014 From: bb010g at gmail.com (Brayden Banks) Date: Tue, 8 Jul 2014 11:39:26 -0700 Subject: [arch-haskell] haskell-lens has out of date deps Message-ID: In the latest update, haskell-lens didn't get fully updated: warning: cannot resolve "haskell-aeson=0.7.0.6-5", a dependency of "haskell-lens" # hackage requirement: 0.7.*; haskell-core version: 0.7.0.6-6 warning: cannot resolve "haskell-bifunctors=4.1.1.1-8", a dependency of "haskell-lens" # hackage requirement: 4.*; haskell-core version: 4.1.1.1-9 warning: cannot resolve "haskell-comonad=4.2-5", a dependency of "haskell-lens" # hackage requirement: 4.*; haskell-core version: 4.2-6 warning: cannot resolve "haskell-free=4.9-1", a dependency of "haskell-lens" # hackage requirement: 4.*; haskell-core version: 4.9-2 warning: cannot resolve "haskell-profunctors=4.0.4-4", a dependency of "haskell-lens" # hackage requirement: 4.*; haskell-core version: 4.0.4-5 warning: cannot resolve "haskell-semigroupoids=4.0.2.1-1", a dependency of "haskell-lens" # hackage requirement: 4.*; haskell-core version: 4.0.2.1-2 warning: cannot resolve "haskell-semigroups=0.15-1", a dependency of "haskell-lens" # hackage requirement: >=0.8.4 && <1; haskell-core version: 0.15.1-1 warning: cannot resolve "haskell-void=0.6.1-65", a dependency of "haskell-lens" # hackage requirement: >=0.5 && <1; haskell-core version: 0.6.1-66 ~Brayden -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Wed Jul 9 05:02:08 2014 From: magnus at therning.org (Magnus Therning) Date: Wed, 9 Jul 2014 07:02:08 +0200 Subject: [arch-haskell] haskell-lens has out of date deps In-Reply-To: References: Message-ID: <20140709050208.GA993@tatooine> On Tue, Jul 08, 2014 at 11:39:26AM -0700, Brayden Banks wrote: > In the latest update, haskell-lens didn't get fully updated: > > warning: cannot resolve "haskell-aeson=0.7.0.6-5", a dependency of > "haskell-lens" > # hackage requirement: 0.7.*; haskell-core version: 0.7.0.6-6 > warning: cannot resolve "haskell-bifunctors=4.1.1.1-8", a dependency of > "haskell-lens" > # hackage requirement: 4.*; haskell-core version: 4.1.1.1-9 > warning: cannot resolve "haskell-comonad=4.2-5", a dependency of > "haskell-lens" > # hackage requirement: 4.*; haskell-core version: 4.2-6 > warning: cannot resolve "haskell-free=4.9-1", a dependency of "haskell-lens" > # hackage requirement: 4.*; haskell-core version: 4.9-2 > warning: cannot resolve "haskell-profunctors=4.0.4-4", a dependency of > "haskell-lens" > # hackage requirement: 4.*; haskell-core version: 4.0.4-5 > warning: cannot resolve "haskell-semigroupoids=4.0.2.1-1", a dependency of > "haskell-lens" > # hackage requirement: 4.*; haskell-core version: 4.0.2.1-2 > warning: cannot resolve "haskell-semigroups=0.15-1", a dependency of > "haskell-lens" > # hackage requirement: >=0.8.4 && <1; haskell-core version: 0.15.1-1 > warning: cannot resolve "haskell-void=0.6.1-65", a dependency of > "haskell-lens" > # hackage requirement: >=0.5 && <1; haskell-core version: 0.6.1-66 There's a very obvious reason for this: ~~~ % pacman -Sl haskell-core|grep lens % ~~~ and ~~~ % cblrepo list|grep -i lens % ~~~ That is, lens is neither in the the HABS database, nor in the repository. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus Heuristic is an algorithm in a clown suit. It?s less predictable, it?s more fun, and it comes without a 30-day, money-back guarantee. -- Steve McConnell, Code Complete -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From magnus at therning.org Wed Jul 9 05:36:40 2014 From: magnus at therning.org (Magnus Therning) Date: Wed, 9 Jul 2014 07:36:40 +0200 Subject: [arch-haskell] haskell-lens has out of date deps In-Reply-To: References: <20140709050208.GA993@tatooine> Message-ID: <20140709053640.GA1743@tatooine> On Tue, Jul 08, 2014 at 10:26:45PM -0700, Brayden Banks wrote: > Sorry, it's in haskell-happstack. Who handles that? From the Arch Haskell Wiki page: https://github.com/tensor5/haskell-happstack/issues /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus I invented the term Object-Oriented, and I can tell you I did not have C++ in mind. -- Alan Kay -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From magnus at therning.org Sun Jul 13 07:32:07 2014 From: magnus at therning.org (Magnus Therning) Date: Sun, 13 Jul 2014 09:32:07 +0200 Subject: [arch-haskell] New GHC -- any wishes on how it should be built? Message-ID: <20140713073207.GA2467@tatooine> Hi all, Now there's a new version of GHC out, 7.8.3. This means a complete rebuild of [haskell-core] is necessary, which also means that NOW is the perfect time to make fundamental changes to our packages, e.g. to change from the 'perf' build profile to 'pref-llvm', or move the location of the documentation. I've not received any such wishes via issues on Github, but maybe there's someone out there thinking "oh, if they only changed this in how Arch Haskell packages are built, then the world would be 10 times better". If you are that someone, then please speak up now! :) I'll start the big rebuild at 12:00 UTC tomorrow, 2014-07-14. So you have until then to voice your suggestions. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus As long as there are ill-defined goals, bizarre bugs, and unrealistic schedules, there will be Real Programmers willing to jump in and Solve The Problem, saving the documentation for later. Long live Fortran! -- Ed Post -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From tensor5 at gmail.com Sun Jul 13 20:20:43 2014 From: tensor5 at gmail.com (Nicola Squartini) Date: Sun, 13 Jul 2014 22:20:43 +0200 Subject: [arch-haskell] haskell-lens has out of date deps In-Reply-To: <20140709053640.GA1743@tatooine> References: <20140709050208.GA993@tatooine> <20140709053640.GA1743@tatooine> Message-ID: I updated [haskell-happstack] just now. Sorry for the delay, but I've been away form my pc for one week. Nicola On Wed, Jul 9, 2014 at 7:36 AM, Magnus Therning wrote: > On Tue, Jul 08, 2014 at 10:26:45PM -0700, Brayden Banks wrote: > > Sorry, it's in haskell-happstack. Who handles that? > > From the Arch Haskell Wiki page: > https://github.com/tensor5/haskell-happstack/issues > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > > I invented the term Object-Oriented, and I can tell you I did not have > C++ in mind. > -- Alan Kay > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://www.haskell.org/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From tensor5 at gmail.com Sun Jul 13 20:35:06 2014 From: tensor5 at gmail.com (Nicola Squartini) Date: Sun, 13 Jul 2014 22:35:06 +0200 Subject: [arch-haskell] New GHC -- any wishes on how it should be built? In-Reply-To: <20140713073207.GA2467@tatooine> References: <20140713073207.GA2467@tatooine> Message-ID: I have some doubts regarding LLVM, since it automatically optimizes the code for the build architecture and chances are that it will not run on older x86_64. I think it's better to wait until llvm-3.5, which should remove the auto optimization. Nicola On Sun, Jul 13, 2014 at 9:32 AM, Magnus Therning wrote: > Hi all, > > Now there's a new version of GHC out, 7.8.3. This means a complete > rebuild of [haskell-core] is necessary, which also means that NOW is > the perfect time to make fundamental changes to our packages, e.g. to > change from the 'perf' build profile to 'pref-llvm', or move the > location of the documentation. > > I've not received any such wishes via issues on Github, but maybe > there's someone out there thinking "oh, if they only changed this in > how Arch Haskell packages are built, then the world would be 10 times > better". If you are that someone, then please speak up now! :) > > I'll start the big rebuild at 12:00 UTC tomorrow, 2014-07-14. So you > have until then to voice your suggestions. > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > > As long as there are ill-defined goals, bizarre bugs, and unrealistic > schedules, there will be Real Programmers willing to jump in and Solve The > Problem, saving the documentation for later. Long live Fortran! > -- Ed Post > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://www.haskell.org/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Mon Jul 14 10:22:56 2014 From: magnus at therning.org (Magnus Therning) Date: Mon, 14 Jul 2014 12:22:56 +0200 Subject: [arch-haskell] New GHC -- any wishes on how it should be built? In-Reply-To: References: <20140713073207.GA2467@tatooine> Message-ID: <20140714102256.GB1953@mtcomp.evidente.local> On Sun, Jul 13, 2014 at 10:35:06PM +0200, Nicola Squartini wrote: > I have some doubts regarding LLVM, since it automatically optimizes > the code for the build architecture and chances are that it will not > run on older x86_64. I think it's better to wait until llvm-3.5, > which should remove the auto optimization. Indeed, turning on LLVM with this version of GHC would require a bit of work on `cblrepo` to make sure the generated PKGBUILDs instructs GHC to tell LLVM that we want "generic" binaries. Complicated sentence, I know, but you know what I mean :) In short, that particular change is unlikely to happen unless there is a killer reason for using LLVM to build. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus As long as there are ill-defined goals, bizarre bugs, and unrealistic schedules, there will be Real Programmers willing to jump in and Solve The Problem, saving the documentation for later. Long live Fortran! -- Ed Post -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From magnus at therning.org Mon Jul 14 10:25:23 2014 From: magnus at therning.org (Magnus Therning) Date: Mon, 14 Jul 2014 12:25:23 +0200 Subject: [arch-haskell] New GHC -- any wishes on how it should be built? In-Reply-To: <20140713073207.GA2467@tatooine> References: <20140713073207.GA2467@tatooine> Message-ID: <20140714102523.GC1953@mtcomp.evidente.local> On Sun, Jul 13, 2014 at 09:32:07AM +0200, Magnus Therning wrote: > Hi all, > > Now there's a new version of GHC out, 7.8.3. This means a complete > rebuild of [haskell-core] is necessary, which also means that NOW is > the perfect time to make fundamental changes to our packages, e.g. > to change from the 'perf' build profile to 'pref-llvm', or move the > location of the documentation. > > I've not received any such wishes via issues on Github, but maybe > there's someone out there thinking "oh, if they only changed this in > how Arch Haskell packages are built, then the world would be 10 > times better". If you are that someone, then please speak up now! > :) > > I'll start the big rebuild at 12:00 UTC tomorrow, 2014-07-14. So > you have until then to voice your suggestions. That's actually a lie! I started the rebuild already yesterday and have a complete repo for you to play with at [haskell-testing] Server = http://xsounds.org/~haskell/testing/$arch Around 12UTC today I'll move it all to [hasell-core], unless someone comes up with something good to change. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus Goto labels should be left-aligned in all caps and should include the programmer's name, home phone number, and credit card number. -- Abdul Nizar -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From magnus at therning.org Fri Jul 18 04:56:36 2014 From: magnus at therning.org (Magnus Therning) Date: Fri, 18 Jul 2014 06:56:36 +0200 Subject: [arch-haskell] [habs] Automatic dependency resolution (#154) In-Reply-To: References: Message-ID: <20140718045636.GA15839@tatooine.lan> Moving this to the mailing list, because this is turning into a discussion. :-) On Thu, Jul 17, 2014 at 02:29:13PM -0700, Profpatsch wrote: > I see. > > I don?t understand how getting a coherent set of package version is > possible manually by hand, though. I?d assume that?s exactly the > kind of things where computers excel. Have you looked at > [stackage](http://www.stackage.org/)? It is possible, but it's a lot of work. Which is exactly why I started working on `cblrepo` in the first place. `cblrepo` does check that dependencies are satisfied, and it reports when they are not. What it doesn't do though is automatically satisfy the missing dependencies. I simply haven't found a need for that. (Patches are welcome, though.) I have looked at stackage. AFAIU it is basically a place to host a subset of Hackage (plus the possibility to patch packages). If I've understood that correctly it means stackage is of VERY limited use to us in maintain Arch Haskell. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus Most software today is very much like an Egyptian pyramid with millions of bricks piled on top of each other, with no structural integrity, but just done by brute force and thousands of slaves. -- Alan Kay -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From mail at profpatsch.de Fri Jul 18 11:49:14 2014 From: mail at profpatsch.de (Profpatsch) Date: Fri, 18 Jul 2014 13:49:14 +0200 Subject: [arch-haskell] Automatic Dependency Resolution Message-ID: <20140718114914.GF10833@nyx> Not a reply, since I just subscribed to this list. > I have looked at stackage. AFAIU it is basically a place to host a > subset of Hackage (plus the possibility to patch packages). If I've > understood that correctly it means stackage is of VERY limited use to > us in maintain Arch Haskell. As far as I understand it their big advantage is that each version is completely coherent, a snapshot of packages that work together. They pipe it through Jenkins and send mails to each package maintainer of packages that don?t build. So they automated the process and accept new packages. Maybe this can be used? e.g. basing Archhaskell on it and just manage the remaining packages manually while slowly integrating them into Stackage? -- Proudly written in Mutt with Vim on Archlinux. Q: Why is this email five sentences or less? A: http://five.sentenc.es -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 819 bytes Desc: not available URL: From alois.cochard at gmail.com Sat Jul 19 12:19:13 2014 From: alois.cochard at gmail.com (Alois Cochard) Date: Sat, 19 Jul 2014 13:19:13 +0100 Subject: [arch-haskell] NodeJS dependency Message-ID: Hi folks, I'm just switching to your awesome [haskell-core] repo (thanks for the effort, this is extremely useful!). But I'm a bit surprised than when re-installing my xmonad... I got a dependency on NodeJS inferred? Is there some form of haskell-webscale integration that I'm not aware of? ;-) Thanks! -- *A\ois* http://twitter.com/aloiscochard http://github.com/aloiscochard -------------- next part -------------- An HTML attachment was scrubbed... URL: From rafal.babinicz at gmail.com Sat Jul 19 18:18:42 2014 From: rafal.babinicz at gmail.com (=?UTF-8?Q?Rafa=C5=82_Babinicz?=) Date: Sat, 19 Jul 2014 20:18:42 +0200 Subject: [arch-haskell] How cblrepo decides when to add 'haskell-' prefix? Message-ID: After cloning habs repo, syncing hackage db with 'cblrepo sync' and adding yeganesh with 'cblrepo add --ghc-version=7.8.3-1 -u https://hackage.haskell.org/package/yeganesh-2.5/yeganesh.cabal' I see different behavior with 'cblrepo pkgbuild taffybar' and 'cblrepo pkgbuild yeganesh'. Former one generates haskell-taffybar with adjusted PKGBUILD for that name, but latter generates regular 'yeganesh' without any haskell-yeganesh. I cannot find any important difference in cblrepo.db and also grepping through source code, I can't find any other references to taffybar than in cblrepo.db which looks similar to yeganesh's entry. Can you clarify in README of cblrepo or habs - how to generate haskell-* package? I've built xmobar and yeganesh with makeahpkg for x86_64 and i686 and I want to contribute, but I'm totally lost here. Should I change names and PKBUILDS by hand? Why cblrepo pkgbuild generates different results for similar packages in cblrepo.db? Thanks in advance for any help. -------------- next part -------------- An HTML attachment was scrubbed... URL: From t at tmh.cc Sat Jul 19 18:43:38 2014 From: t at tmh.cc (Taylor Hedberg) Date: Sat, 19 Jul 2014 11:43:38 -0700 Subject: [arch-haskell] How cblrepo decides when to add 'haskell-' prefix? In-Reply-To: References: Message-ID: I believe (though someone else may correct me here), that the "haskell-" is only prepended to the package name if the package has a library component. If the package comprises only executables, as in the case of yeganesh, then it is left as-is. On Sat, Jul 19, 2014 at 11:18 AM, Rafa? Babinicz wrote: > After cloning habs repo, > syncing hackage db with 'cblrepo sync' > and adding yeganesh with 'cblrepo add --ghc-version=7.8.3-1 -u > https://hackage.haskell.org/package/yeganesh-2.5/yeganesh.cabal' > I see different behavior with 'cblrepo pkgbuild taffybar' and > 'cblrepo pkgbuild yeganesh'. Former one generates > haskell-taffybar with adjusted PKGBUILD for that name, but latter > generates regular 'yeganesh' without any haskell-yeganesh. > > I cannot find any important difference in cblrepo.db and also > grepping through source code, I can't find any other references to > taffybar than in cblrepo.db which looks similar to yeganesh's entry. > > Can you clarify in README of cblrepo or habs - how to generate > haskell-* package? > > I've built xmobar and yeganesh with makeahpkg for x86_64 and i686 > and I want to contribute, but I'm totally lost here. > > Should I change names and PKBUILDS by hand? Why cblrepo pkgbuild > generates different results for similar packages in cblrepo.db? > > Thanks in advance for any help. > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://www.haskell.org/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Sat Jul 19 20:00:53 2014 From: magnus at therning.org (Magnus Therning) Date: Sat, 19 Jul 2014 22:00:53 +0200 Subject: [arch-haskell] NodeJS dependency In-Reply-To: References: Message-ID: <20140719200053.GA1436@tatooine.lan> On Sat, Jul 19, 2014 at 01:19:13PM +0100, Alois Cochard wrote: > Hi folks, > > I'm just switching to your awesome [haskell-core] repo (thanks for > the effort, this is extremely useful!). > > But I'm a bit surprised than when re-installing my xmonad... I got a > dependency on NodeJS inferred? > > Is there some form of haskell-webscale integration that I'm not > aware of? ;-) Nope, no that I'm aware of. Please report the reason for it if you track it down :) /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus As long as there are ill-defined goals, bizarre bugs, and unrealistic schedules, there will be Real Programmers willing to jump in and Solve The Problem, saving the documentation for later. Long live Fortran! -- Ed Post -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From magnus at therning.org Sat Jul 19 20:01:32 2014 From: magnus at therning.org (Magnus Therning) Date: Sat, 19 Jul 2014 22:01:32 +0200 Subject: [arch-haskell] How cblrepo decides when to add 'haskell-' prefix? In-Reply-To: References: Message-ID: <20140719200132.GB1436@tatooine.lan> On Sat, Jul 19, 2014 at 11:43:38AM -0700, Taylor Hedberg wrote: > I believe (though someone else may correct me here), that the > "haskell-" is only prepended to the package name if the package has > a library component. If the package comprises only executables, as > in the case of yeganesh, then it is left as-is. That is exactly it! /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus Most software today is very much like an Egyptian pyramid with millions of bricks piled on top of each other, with no structural integrity, but just done by brute force and thousands of slaves. -- Alan Kay -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From alois.cochard at gmail.com Sat Jul 19 20:54:05 2014 From: alois.cochard at gmail.com (Alois Cochard) Date: Sat, 19 Jul 2014 21:54:05 +0100 Subject: [arch-haskell] NodeJS dependency In-Reply-To: <20140719200053.GA1436@tatooine.lan> References: <20140719200053.GA1436@tatooine.lan> Message-ID: My bad, I was very surprised to see nodejs when doing a '-S haskell-xmonad'. But I suppose it's because I did '-yy' at some point before and there was a new version in the cache.... because indeed, trying to track down there is nothing. The worst thing is that I exposed the fact that I have a dependency on nodejs in my system ;-) Sorry for the noise On 19 July 2014 21:00, Magnus Therning wrote: > On Sat, Jul 19, 2014 at 01:19:13PM +0100, Alois Cochard wrote: > > Hi folks, > > > > I'm just switching to your awesome [haskell-core] repo (thanks for > > the effort, this is extremely useful!). > > > > But I'm a bit surprised than when re-installing my xmonad... I got a > > dependency on NodeJS inferred? > > > > Is there some form of haskell-webscale integration that I'm not > > aware of? ;-) > > Nope, no that I'm aware of. Please report the reason for it if you > track it down :) > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > > As long as there are ill-defined goals, bizarre bugs, and unrealistic > schedules, there will be Real Programmers willing to jump in and Solve The > Problem, saving the documentation for later. Long live Fortran! > -- Ed Post > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://www.haskell.org/mailman/listinfo/arch-haskell > > -- *A\ois* http://twitter.com/aloiscochard http://github.com/aloiscochard -------------- next part -------------- An HTML attachment was scrubbed... URL: From rudy at matela.com.br Mon Jul 28 16:07:34 2014 From: rudy at matela.com.br (Rudy Matela) Date: Mon, 28 Jul 2014 17:07:34 +0100 Subject: [arch-haskell] Bootstrapping a cblrepo.db / custom repository Message-ID: Hello All, First of all, many thanks to Magnus Therning for maintaining the arch-haskell repository and it's tools. It saved a lot of time for me (both fighting cabal and compiling packages). I'm planning to use cblrepo to create a new repository from scratch. So, I have a question: How to use cblrepo to bootstrap a new repository? (Is it even intended to use for that purpose.) Or, more broadly, what's the best way to bootstrap an arch (-haskell) repository? For example, suppose I want a repository with just template-haskell: $ mkdir habs-bootstrap $ cd habs-bootstrap $ cblrepo add template-haskell,2.9.0.0 Failed to satisfy the following dependencies for template-haskell: base ==4.7.* containers ==0.5.* pretty ==1.1.* Shouldn't the dependencies here be added by default? Guessing here that it is not possible because of inability to guess package types (--ghc-pkg / --distro-pkg). I've also tried: $ cblrepo add -g base,4.7.0.1 That was ok, but: $ cblrepo add template-haskell,2.9.0.0 Failed to satisfy the following dependencies for template-haskell: containers ==0.5.* pretty ==1.1.* Containers and pretty are provided by GHC. Should I add GHC in my repository? That does not seem to be the case with haskell-happstack [1]: it has array on it's db (it is also provided by ghc). Regards, Rudy [1]: https://github.com/tensor5/haskell-happstack PS: Which kind of setup do you all use, in general, to maintain your own packages not provided by haskell-core? So far I've been using my own fork of archhaskell/habs, but I kinda loose the work (on cblrepo.db) when merging with Magnus' branch, needing to re-add all my packages. From tensor5 at gmail.com Mon Jul 28 18:44:22 2014 From: tensor5 at gmail.com (Nicola Squartini) Date: Mon, 28 Jul 2014 20:44:22 +0200 Subject: [arch-haskell] Bootstrapping a cblrepo.db / custom repository In-Reply-To: References: Message-ID: You have to add each dependency individually, even though they are all provided by GHC. In your example: $ cblrepo add -g base,4.7.0.1 -g containers,0.5.5.1 -g pretty,1.1.1.1 template-haskell,2.9.0.0 On Mon, Jul 28, 2014 at 6:07 PM, Rudy Matela wrote: > Hello All, > > First of all, many thanks to Magnus Therning for maintaining the > arch-haskell repository and it's tools. It saved a lot of time for me > (both fighting cabal and compiling packages). > > I'm planning to use cblrepo to create a new repository from scratch. > So, I have a question: > > How to use cblrepo to bootstrap a new repository? (Is it even > intended to use for that purpose.) Or, more broadly, what's the best > way to bootstrap an arch (-haskell) repository? > > For example, suppose I want a repository with just template-haskell: > > $ mkdir habs-bootstrap > $ cd habs-bootstrap > $ cblrepo add template-haskell,2.9.0.0 > Failed to satisfy the following dependencies for template-haskell: > base ==4.7.* > containers ==0.5.* > pretty ==1.1.* > > Shouldn't the dependencies here be added by default? Guessing here > that it is not possible because of inability to guess package types > (--ghc-pkg / --distro-pkg). > > I've also tried: > > $ cblrepo add -g base,4.7.0.1 > > That was ok, but: > > $ cblrepo add template-haskell,2.9.0.0 > Failed to satisfy the following dependencies for template-haskell: > containers ==0.5.* > pretty ==1.1.* > > Containers and pretty are provided by GHC. Should I add GHC in my > repository? That does not seem to be the case with haskell-happstack > [1]: it has array on it's db (it is also provided by ghc). > > Regards, > Rudy > > > [1]: https://github.com/tensor5/haskell-happstack > > > PS: Which kind of setup do you all use, in general, to maintain your > own packages not provided by haskell-core? So far I've been using my > own fork of archhaskell/habs, but I kinda loose the work (on > cblrepo.db) when merging with Magnus' branch, needing to re-add all my > packages. > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://www.haskell.org/mailman/listinfo/arch-haskell > -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Tue Jul 29 07:50:26 2014 From: magnus at therning.org (Magnus Therning) Date: Tue, 29 Jul 2014 09:50:26 +0200 Subject: [arch-haskell] Bootstrapping a cblrepo.db / custom repository In-Reply-To: References: Message-ID: <20140729075026.GA8103@mtcomp.evidente.local> On Mon, Jul 28, 2014 at 08:44:22PM +0200, Nicola Squartini wrote: > You have to add each dependency individually, even though they are all > provided by GHC. In your example: > > $ cblrepo add -g base,4.7.0.1 -g containers,0.5.5.1 > -g pretty,1.1.1.1 template-haskell,2.9.0.0 Indeed you have to manually add all the basic modules. In the cblrepo github repo you can find a list of packages for 7.8.2[1]. I still have to add one for 7.8.3. Another option is to look at the database in habs. Either you run `cblrepo -g --no-repo` and pipe it to some suitable sed/awk script, or you just copy the first lines from the database (the database has three sections, of which we only use 2, the first one contains the ones added with -g). Hope this helps. /M [1]: https://github.com/magthe/cblrepo/blob/master/data/ghc-7.8.2_pkgs -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus Code as if whoever maintains your program is a violent psychopath who knows where you live. -- Anonymous -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From rudy at matela.com.br Tue Jul 29 11:30:23 2014 From: rudy at matela.com.br (Rudy Matela) Date: Tue, 29 Jul 2014 12:30:23 +0100 Subject: [arch-haskell] Bootstrapping a cblrepo.db / custom repository In-Reply-To: <20140729075026.GA8103@mtcomp.evidente.local> References: <20140729075026.GA8103@mtcomp.evidente.local> Message-ID: On Tue, Jul 29, 2014 at 8:50 AM, Magnus Therning wrote: > > On Mon, Jul 28, 2014 at 08:44:22PM +0200, Nicola Squartini wrote: > > You have to add each dependency individually, even though they are all > > provided by GHC. In your example: > > > > $ cblrepo add -g base,4.7.0.1 -g containers,0.5.5.1 > > -g pretty,1.1.1.1 template-haskell,2.9.0.0 > > Indeed you have to manually add all the basic modules. > > In the cblrepo github repo you can find a list of packages for > 7.8.2[1]. I still have to add one for 7.8.3. > > Another option is to look at the database in habs. Either you run > `cblrepo -g --no-repo` and pipe it to some suitable sed/awk script, or > you just copy the first lines from the database (the database has > three sections, of which we only use 2, the first one contains the > ones added with -g). Thanks Nicola and Magnus, I'll try those options and see which works best! From tensor5 at gmail.com Wed Jul 30 13:30:55 2014 From: tensor5 at gmail.com (Nicola Squartini) Date: Wed, 30 Jul 2014 15:30:55 +0200 Subject: [arch-haskell] Issue with i686 packages Message-ID: Hi Magnus, I'm trying to build updated packages for [haskell-happstack] and I get this kind of errors: $ ./makeahpkg -c -a i686 $( Making package: haskell-fb 1.0.2-6 (Wed Jul 30 15:18:26 CEST 2014) ==> Retrieving sources... -> Found fb-1.0.2.tar.gz ==> Validating source files with sha256sums... fb-1.0.2.tar.gz ... Passed ==> Making package: haskell-fb 1.0.2-6 (Wed Jul 30 15:18:27 CEST 2014) ==> Checking runtime dependencies... ==> Installing missing dependencies... error: target not found: haskell-conduit-extra=1.1.3-1 error: target not found: haskell-monad-logger=0.3.7-4 ==> ERROR: 'pacman' failed to install missing dependencies. ==> ERROR: Build failed, check /home/nicola/haskell-happstack/i686-chroot/build/build [...] and similar others for other packages. The build process ran flawless for x86_64. I assume there is some problem with the i686 packages after today's update of habs. Thanks Nicola -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Wed Jul 30 13:56:17 2014 From: magnus at therning.org (Magnus Therning) Date: Wed, 30 Jul 2014 15:56:17 +0200 Subject: [arch-haskell] Issue with i686 packages In-Reply-To: References: Message-ID: <20140730135617.GA9752@mtcomp.evidente.local> On Wed, Jul 30, 2014 at 03:30:55PM +0200, Nicola Squartini wrote: > Hi Magnus, > > I'm trying to build updated packages for [haskell-happstack] and I get this > kind of errors: > > $ ./makeahpkg -c -a i686 $( > [...] > ==> Making package: haskell-fb 1.0.2-6 (Wed Jul 30 15:18:26 CEST 2014) > ==> Retrieving sources... > -> Found fb-1.0.2.tar.gz > ==> Validating source files with sha256sums... > fb-1.0.2.tar.gz ... Passed > ==> Making package: haskell-fb 1.0.2-6 (Wed Jul 30 15:18:27 CEST 2014) > ==> Checking runtime dependencies... > ==> Installing missing dependencies... > error: target not found: haskell-conduit-extra=1.1.3-1 > error: target not found: haskell-monad-logger=0.3.7-4 > ==> ERROR: 'pacman' failed to install missing dependencies. > > ==> ERROR: Build failed, check > /home/nicola/haskell-happstack/i686-chroot/build/build > [...] > > and similar others for other packages. The build process ran > flawless for x86_64. I assume there is some problem with the i686 > packages after today's update of habs. Yes, that sounds likely. I'm not sure when I'll time to look into it though. Do you mind looking into it a bit and list the affected packages? /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus Heuristic is an algorithm in a clown suit. It?s less predictable, it?s more fun, and it comes without a 30-day, money-back guarantee. -- Steve McConnell, Code Complete -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From tensor5 at gmail.com Wed Jul 30 17:13:14 2014 From: tensor5 at gmail.com (Nicola Squartini) Date: Wed, 30 Jul 2014 19:13:14 +0200 Subject: [arch-haskell] Issue with i686 packages In-Reply-To: <20140730135617.GA9752@mtcomp.evidente.local> References: <20140730135617.GA9752@mtcomp.evidente.local> Message-ID: I gathered some: haskell-bifunctors=4.1.1.1-13 haskell-conduit=1.1.7-3 haskell-conduit-extra=1.1.3-1 haskell-contravariant=1.1-1 haskell-free=4.9-6 haskell-monad-logger=0.3.7-4 haskell-pandoc=1.12.4.2-12 haskell-profunctors=4.2-1 haskell-semigroupoids=4.0.4-1 haskell-semigroups=0.15.2-1 haskell-void=0.6.1-70 haskell-xml-conduit=1.2.1-3 They are those updated on 2014-07-30. Maybe the haskell-core.db needs to be regenerated. However now the error message is slightly different: $ ./makeahpkg -c -a i686 $( Making package: haskell-fb 1.0.2-6 (Wed Jul 30 19:08:31 CEST 2014) ==> Retrieving sources... -> Found fb-1.0.2.tar.gz ==> Validating source files with sha256sums... fb-1.0.2.tar.gz ... Passed ==> Making package: haskell-fb 1.0.2-6 (Wed Jul 30 19:08:31 CEST 2014) ==> Checking runtime dependencies... ==> Installing missing dependencies... resolving dependencies... warning: cannot resolve "haskell-void=0.6.1-70", a dependency of "haskell-conduit" warning: cannot resolve "haskell-void=0.6.1-70", a dependency of "haskell-conduit" warning: cannot resolve "haskell-conduit=1.1.7-3", a dependency of "haskell-conduit-extra" warning: cannot resolve "haskell-void=0.6.1-70", a dependency of "haskell-conduit" warning: cannot resolve "haskell-conduit=1.1.7-3", a dependency of "haskell-http-conduit" warning: cannot resolve "haskell-void=0.6.1-70", a dependency of "haskell-conduit" warning: cannot resolve "haskell-conduit=1.1.7-3", a dependency of "haskell-monad-logger" warning: cannot resolve "haskell-void=0.6.1-70", a dependency of "haskell-conduit" warning: cannot resolve "haskell-conduit=1.1.7-3", a dependency of "haskell-conduit-extra" warning: cannot resolve "haskell-conduit-extra=1.1.3-1", a dependency of "haskell-monad-logger" :: The following packages cannot be upgraded due to unresolvable dependencies: haskell-conduit haskell-conduit-extra haskell-http-conduit haskell-monad-logger :: Do you want to skip the above packages for this upgrade? [y/N] error: failed to prepare transaction (unexpected error) ==> ERROR: 'pacman' failed to install missing dependencies. ==> ERROR: Build failed, check /home/nicola/haskell-happstack/i686-chroot/build/build [...] Nicola On Wed, Jul 30, 2014 at 3:56 PM, Magnus Therning wrote: > On Wed, Jul 30, 2014 at 03:30:55PM +0200, Nicola Squartini wrote: > > Hi Magnus, > > > > I'm trying to build updated packages for [haskell-happstack] and I get > this > > kind of errors: > > > > $ ./makeahpkg -c -a i686 $( > > > [...] > > ==> Making package: haskell-fb 1.0.2-6 (Wed Jul 30 15:18:26 CEST 2014) > > ==> Retrieving sources... > > -> Found fb-1.0.2.tar.gz > > ==> Validating source files with sha256sums... > > fb-1.0.2.tar.gz ... Passed > > ==> Making package: haskell-fb 1.0.2-6 (Wed Jul 30 15:18:27 CEST 2014) > > ==> Checking runtime dependencies... > > ==> Installing missing dependencies... > > error: target not found: haskell-conduit-extra=1.1.3-1 > > error: target not found: haskell-monad-logger=0.3.7-4 > > ==> ERROR: 'pacman' failed to install missing dependencies. > > > > ==> ERROR: Build failed, check > > /home/nicola/haskell-happstack/i686-chroot/build/build > > [...] > > > > and similar others for other packages. The build process ran > > flawless for x86_64. I assume there is some problem with the i686 > > packages after today's update of habs. > > Yes, that sounds likely. > > I'm not sure when I'll time to look into it though. Do you mind > looking into it a bit and list the affected packages? > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > > Heuristic is an algorithm in a clown suit. It?s less predictable, it?s more > fun, and it comes without a 30-day, money-back guarantee. > -- Steve McConnell, Code Complete > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://www.haskell.org/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From magnus at therning.org Thu Jul 31 04:18:27 2014 From: magnus at therning.org (Magnus Therning) Date: Thu, 31 Jul 2014 06:18:27 +0200 Subject: [arch-haskell] Issue with i686 packages In-Reply-To: References: <20140730135617.GA9752@mtcomp.evidente.local> Message-ID: <20140731041827.GA1546@tatooine.lan> On Wed, Jul 30, 2014 at 07:13:14PM +0200, Nicola Squartini wrote: > I gathered some: > > haskell-bifunctors=4.1.1.1-13 > haskell-conduit=1.1.7-3 > haskell-conduit-extra=1.1.3-1 [...] Thanks. Make another attempt now, hopefully it'll work a bit better now. Sorry for the mess. /M -- Magnus Therning OpenPGP: 0xAB4DFBA4 email: magnus at therning.org jabber: magnus at therning.org twitter: magthe http://therning.org/magnus In a hierarchy, every employee tends to rise to his level of incompetence. -- The Peter Principle -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 181 bytes Desc: not available URL: From tensor5 at gmail.com Thu Jul 31 10:04:59 2014 From: tensor5 at gmail.com (Nicola Squartini) Date: Thu, 31 Jul 2014 12:04:59 +0200 Subject: [arch-haskell] Issue with i686 packages In-Reply-To: <20140731041827.GA1546@tatooine.lan> References: <20140730135617.GA9752@mtcomp.evidente.local> <20140731041827.GA1546@tatooine.lan> Message-ID: It works now! Thanks Nicola On Thu, Jul 31, 2014 at 6:18 AM, Magnus Therning wrote: > On Wed, Jul 30, 2014 at 07:13:14PM +0200, Nicola Squartini wrote: > > I gathered some: > > > > haskell-bifunctors=4.1.1.1-13 > > haskell-conduit=1.1.7-3 > > haskell-conduit-extra=1.1.3-1 > [...] > > Thanks. Make another attempt now, hopefully it'll work a bit better > now. Sorry for the mess. > > /M > > -- > Magnus Therning OpenPGP: 0xAB4DFBA4 > email: magnus at therning.org jabber: magnus at therning.org > twitter: magthe http://therning.org/magnus > > In a hierarchy, every employee tends to rise to his level of incompetence. > -- The Peter Principle > > _______________________________________________ > arch-haskell mailing list > arch-haskell at haskell.org > http://www.haskell.org/mailman/listinfo/arch-haskell > > -------------- next part -------------- An HTML attachment was scrubbed... URL: